Call Analytics for Installers: Judge Leads by the Call
Call analytics lets installers judge lead quality from what was said on the call, not from memory. What it should show, and how to use it fairly.
Call analytics for installers means judging a lead by what was actually said on the call, rather than by someone's memory of it. Most arguments about lead quality run on recollection on both sides: the installer says the leads were rubbish, the supplier says the team didn't ring them properly. The call is the evidence that settles it.
That evidence cuts both ways. Sometimes it shows a lead that should never have been sent. Sometimes it shows a good lead your team let slip. Both are worth knowing.
Why do lead-quality arguments go nowhere?
Because neither side is looking at the same thing.
The sales manager at a solar installer in Warrington remembers Friday afternoon: three calls in a row to people who didn't seem to want anything. Tuesday's survey booking is less vivid. The supplier, meanwhile, sees a status in a CRM, "not interested", with nothing behind it.
A status records what happened, not why. "Not interested" can mean a homeowner who never asked to be contacted. It can also mean a homeowner who asked for a rough price three times, didn't get one, and gave up. Those need opposite fixes, and the status alone can't tell them apart.
What should call analytics give an installer?
Good call recording analysis does four jobs. If a tool gives you only a recording and a transcript, you still have most of the work to do yourself.
1. The moments that mattered
Nobody has time to listen to every seven-minute call. The call should be broken into its moments, so you can go straight to the part you need: where the homeowner asked about a grant, confirmed they own the home, asked for a battery price, raised a worry about cost or booked the survey.
2. What the customer said against what the form said
The form is the homeowner's answer, given on a phone in a minute or two. The call is a longer conversation. Putting them side by side shows where they differ: the form said £100–£200 a month on electricity, the call said about £90.
Most differences are details, and they're normal. Some are failed checks. Our guide to lead qualification questions explains where to draw that line, and which questions belong on the form at all.
3. The outcome, logged
Every call should end with an outcome from a short, fixed list: booked, call back, not interested, invalid number. Use the same names every time, so a week of outcomes can be counted and compared.
Those outcomes are what the ads should learn from. Recorded consistently, they can be sent back to the campaigns, which is the idea behind a lead feedback loop.
4. Coaching for the rep
A short note per call on what worked, what to try and what to ask next time. More on doing that well below.
A note on terms. Call tracking for installers usually means tracking which ad or phone number a call came from. That's useful for knowing where calls originate, but it tells you nothing about what was said. Call analytics answers the second question.
How do you tell a poor lead from a good one that slipped?
Look at what the call shows, and ask who could have prevented the result.
| What the call shows | What it points to | Who should act |
|---|---|---|
| The home is outside your postcodes | A poor lead: it failed a basic check | The supplier |
| The person rents, or is calling on behalf of the owner | A poor lead: not the homeowner | The supplier |
| The homeowner says they never asked to be contacted, and the consent record can't show otherwise | A poor lead | The supplier |
| The homeowner asked for a rough price twice and didn't get one | A good lead that slipped | Your team |
| A survey was booked without asking who else decides, and the partner cancelled it | A good lead that slipped | Your team |
| The bill is nearer £90 than the £100–£200 on the form | Neither: a normal gap between form and call | Note it before the survey |
The third row needs care. A homeowner who has forgotten a form they filled in that morning is not the same as one who never filled it in. The consent's stored wording and time will usually tell you which.
Be honest, too, about what a recording can't show. One call proves little either way, so look for a pattern across weeks. A number that rang out has no recording, which makes "no answer, three tries" a status rather than evidence about the lead. And a recording shows what was said, not what the homeowner decided afterwards.
How do you use call evidence fairly with a supplier?
Call evidence makes a conversation about lead quality shorter and calmer, provided both sides agree the rules first:
- agree up front what counts as a poor lead (outside the area, not the homeowner, a dead number, no consent naming you) and what doesn't (a detail that differs from the form, a homeowner who isn't ready yet);
- send the moment, not a summary: "at 1:50 she says the house belongs to her parents" is a fact, while "the leads are poor" is an opinion;
- judge the pattern over several weeks, not one bad afternoon;
- accept the calls that show your side let a lead slip;
- know the supplier's returns policy, and what it covers, before you start.
Whether you can see what happened to a lead after delivery is one of the questions worth asking before you buy. It's covered in our guide to how to choose a lead generation company.
How do you coach from calls without micromanaging?
Sales call coaching goes wrong when every call gets a score and every sentence gets a comment. Reps start performing for the recording instead of talking to the homeowner.
A lighter pattern is easier to act on: one thing that worked, one thing to try and one thing to ask next time. On an example solar call, that might read:
- Worked: the rep answered the grant question straight away and moved on, and the homeowner stayed with him.
- Try: give a ballpark battery price before the survey. She asked twice.
- Ask next time: ask about electric car plans early, because they change the battery size.
That's three lines a rep can use on tomorrow's calls. A few habits help it stick:
- review a handful of calls a week together, not all of them;
- let the rep choose one of the calls;
- start with what worked, and be specific;
- coach the question, not the person: "ask who else decides before you talk price" is easier to hear than "you rushed it".
One piece of good practice applies whatever tool you use: tell callers that calls are recorded, and cover recordings in your privacy notice.
Where POCKLA fits
POCKLA Desk includes free call analytics on calls made from Desk. Each call is broken into its moments and compared with the homeowner's form, so "£100–£200 a month" on the form and "about £90" on the call show up side by side. Every call gets coaching on what worked, what to try and what to ask next time, and the outcome is logged: booked, call back, not interested or invalid number.
If you don't have a CRM, you can work your leads in Desk: call from Desk on POCKLA's line, log outcomes and set callbacks. If you do have one, leads go into it with every answer attached and a timeline from form to follow-up to delivery.
Desk is one part of how POCKLA works: campaigns reach homeowners, each audience lands on a page made for it, AI follows up on WhatsApp or SMS in your business's name, and the outcome, logged in Desk or read from your CRM, goes back to the ads. Leads are exclusive, sold once to one buyer, and for solar, invalid numbers are returned with no time limit.
When a lead is poor, the call shows why. When a good one slips, you'll see that too.
Want to see what happens on every call? See POCKLA Desk →